On today's episode we talk to Larry Allen and Dan Matlock about decision making around destination therapy. No this has nothing to do with your summer vacation plans. Rather, we talk about how individuals with heart failure decide about whether or not to pursue "destination therapy" with an Left Ventricular Device, or LVAD.
Larry Allen is a Transplant Cardiologist at the University of Colorado and Dan Matlock is a General Internist, a Geriatrician, and Palliative Care doctor at the University of Colorado. We talk with them about their recent JAMA Internal Medicine article “The Effectiveness of an Intervention Supporting Shared Decision Making for Destination Therapy Left Ventricular Assist Device” or the DECIDE-LVAD Randomized Control Trial. We also discuss other topics including their pretty amazing (and free) website called patientdecisionaid.org which includes not only the LVAD decision aid, but there's also decision aids on ICD's, cardiac resynchronization therapy and defibrillation, and one for colon cancer.
Listen to GeriPal Podcasts on:
Eric: Welcome to the GeriPal Podcast, this is Eric Widera.
Alex: This is Alex Smith.
Eric: And Alex, who are our guests today?
Alex: Today our guests are Larry Allen, who is a Transplant Cardiologist at the University of Colorado. Welcome to the GeriPal Podcast, Larry.
Larry: Thank you.
Alex: And our other guest is a recurring part of the GeriPal Podcast. I think you've been a host at least, if not a guest, in the past. Dan Matlock, who's a General Internist, a Geriatrician, and Palliative Care doc at the University of Colorado. Welcome back to the GeriPal Podcast, Dan.
Dan: Thank you.
Eric: Welcome, Dan. Dan, you know the drill. Song request for Alex.
Dan: I think one of my favorite songs is While My Guitar Gently Weeps by The Beetles. So I think I'd like to request that.
Eric: All right. Alex.
Alex: All right, that's enough for now. Well I got most of the lyrics, but the alternative was what? My Milkshake and the boys?
Eric: My milkshake brings all the boys to the yard. Is that right, Dan?
Dan: Yeah, I would've loved to have heard that as well. Or Iron Maiden's The Number of the Beast, those would have also been acceptable alternatives.
Eric: What's the theme there, Dan, what's the theme?
Dan: I'm not sure there is one.
Eric: To challenge Alex is the theme.
Alex: Eric was rooting for My Milkshake brings all the boys to the yard.
Eric: I strongly objected to My Guitar Gently Weeps. I wanted My Milkshake.
Alex: And I loved While My Guitar Gently Weeps because I've always wanted to learn that song. And I got some of the lyrics right, but we'll see if I can do better at the end.
Eric: We gotta spice it up for the younger audience, Alex. No more of like '60s and '70s songs.
So on today's episode we were hoping to talk to you both about your JAMA Internal Medicine article “The Effectiveness of an Intervention Supporting Shared Decision Making for Destination Therapy Left Ventricular Assist Device” or the DECIDE-LVAD Randomized Control Trial.
So maybe before we go deep into the subject, can you describe a little bit, both of you, kind of how you got interested in this subject?
Dan: Well, this is Dan, I'll start. I got interested in shared decision making generally just as a resident with experiences with patients who seemed to be getting things and offered things and apparently had no idea what they were getting into. I initially got into defibrillators and then I was lucky enough Larry Allen moved to Colorado and he and I started collaborating in 2008, 2009. And it's been just a really wonderful collaboration so that he pulled me into the LVAD space with the shared decision making work.
Larry: Yeah and this is Larry. I kind of came at it from a different angle. I've always been interested in patients who are living with chronic heart failure who then start to fail more typical therapies like drugs or even pacing devices and then have to either decide when they're appropriate for hospice or whether if they're a candidate to pursue major therapies, like transplants or artificial heart technology. So I started out trying to predict that and then realized that the real interesting piece of that was actually trying to grapple with the uncertainty around it and how patients have to make these tough decisions about options that aren't necessarily what they want either way. Which is to pursue hospice, which may not be their first choice, but yet to pursue an LVAD which can be life saving but can also be really burdensome, expensive, and doesn't always have great outcomes.
So that's how I got at it and Dan was really an expert in our institution thinking about how patients deal with those kind of tough decisions.
Alex: You know I think it'd be helpful for some our listeners who may or may not be familiar with what an LVAD is and what destination therapy is, to talk, just briefly introduce what is an LVAD and what is destination therapy mean.
Larry: So there are about six million people in the U.S. who have heart failure and each year approximately maybe 50 to 100,000 of those patients will progress to essentially dying from their heart failure. And those patients are really left with the option of dying from their heart failure or pursuing kind of heart replacement therapies.
The issue for heart replacement therapy is that transplants are fairly limited. There's about 3,000 suitable organs in the United States every year. And then the only other alternative is durable or permanent artificial heart technology. And the current technologies have gotten a lot better, but they still have some limitations and some design characteristics. So they were first kind of designed just as short term therapy to get people to transplant, but now the current devices can last five, 10, or even 15 years longer.
But the other problem with them is that they, really the ones that work well are just plugged into the left ventricle and sit beside the heart. So the patients really have to have left greater than right sided heart failure and they have to be able to take anticoagulations. So there's a certain percentage of patients who may not be transplant eligible who are dying of their heart failure who might be a candidate for a left ventricular assist device. And if they're one of these people who can't get a transplant, but who could potentially get a VAD, then if they proceed with that they would get a surgical implant of an LVAD and live with that the device for the rest of their life, typically until they die. And we call that destination therapy.
Alex: So the destination is ...
Larry: Rather than the destination being transplant, the destination is the device itself and ultimately death.
Alex: So we talked about what an LVAD is and what destination therapy is, and talked sort of about the scope of the problem. Tell us what you did here in this study.
Dan: So this is Dan. I'll summarize it here. We spent several years, even before this study making a decision aid. We made an eight page decision aid and about a 25 minute video. We went through 28 versions of the paper tool and five versions of the video tool testing this with lots of patients and clinicians to try and get it so it was balanced, that is it wasn't trying to encourage or discourage people from getting one of these devices. Trying to make sure it was understandable and readable and accurate. And that's really the process of developing the intervention. And that was several years of work, including a lot of qualitative interviews to even understand what patients needed to know.
Once we developed that, then we applied for this grant from PCORI to study this in six programs around the country. And we did what we called a combined effectiveness and implementation study, where we wanted to look and see if this decision aid was effective and we wanted to understand actually how do you implement this in the real world practice. So we had sort of this interesting step-wedge design, which allows us to look at both effectiveness and implementation.
I mean in terms of effectiveness we were trying to see if we could help patients understand things better. So we measured knowledge and we measured if the treatment they chose was actually concordant with their values.
Alex: And just backing up a little bit to the intervention itself and the decision aid itself. Could you tell us a little bit more about what was involved in the decision aid, how much time it took, and how much effort was required on the part of the participant or other folks?
Dan: So we ... Most of these patients were in the hospital, so they're a bit of a captive audience. It would probably take somebody a half hour and an hour to go through this. We did not allow our study staff to give the decision aid out. That's part of the implementation design. We instead chose to try and encourage the sites and the clinicians at the sites to develop a process to give the decision aid out.
And each site did that a little bit differently. One of the things we studied was how did one site do it differently than another site. The only rule we had was they had to give it out before they made the decision, before they had the discussion with them. So it sort of depended. Some people, some sites weren't able to use the video and only used the paper tool, at least one site was like that. Other sites would use the video and bring a TV into the room. Other sites would use it on the phone. So it was a little bit different for each patient at each site, depending on how each site chose to implement the tool.
Alex: And the pamphlet is available in the supplement to the publication in JAMA Internal Medicine. Is the video also available for folks to see?
Dan: Yeah, we have them on our website. And I'll plug it, www.patientdecisionaid.org, they are available and free to use. We have a terms on there that people can look at. We want people to use these things for clinical work. We want to protect the intellectual property for our own research, but other than that, we want people to use it. So it's out there.
Larry: Yeah, you know Dan and our group, we've actually done work in other areas too. So that patient decisionaid.org website includes not only the LVAD decision aid, but there's an ICD decision aid and other decisions around EP devices for heart failure. And we also have a decision aid for sacubitril valsartan for heart failure. There's a colonoscopy screening decision aid there. And atrial fibrillation stroke prevention as well.
So there's a couple of different tools, we'd love for people to use them for their patients.
Eric: And what's the link again?
Eric: Awesome. We'll have that link on our GeriPal website associated with this podcast as well. Maybe we can actually spend just a minute talking about, so you talked about some of the things that you were trying to achieve in this study. What did you actually achieve? Like what did this ... How did this decision aid compare to the control group?
Larry: Well I'll go with that. Dan kind of highlighted that this was a combination effectiveness and implementation trial. So for the effectiveness we wanted to see whether the decision aids when given to patients and their caregivers by the LVAD or mechanical circulatory support team, whether those patients decision aids actually helped patients. And we looked at decision quality. So are patients making good decisions about whether they should or should not get an LVAD given their own values, preferences, and goals in the context of their medical realities.
And we found that it did. So patients were more knowledgeable about the decision they were making, including the LVAD device and the alternatives in the period of the study where they were getting this decision aid compared to the usual care or marketing materials.
And we also found that the patients who got this decision aid were more likely to choose LVAD or not consistent with whether they really wanted to be aggressive with their care and were willing to undergo a major surgery and be hooked up to an electrical cord and take care of a machine, versus rather they didn't want to do that and wanted to kind of live what time they had left with a little less interface with the medical community and less risk taking. So we found that people who were aggressive were more likely to say they wanted a VAD and the patients who were kind of not interested in that were actually more likely to turn it down. So we consider that important.
A secondary end point we looked at was how often did patients go on to get an LVAD. And we found that actually once we implemented the patient decision aid, that patients look like they were less likely to go on to that. So we were before implementing the decision aid about 70 odd percent of patients went to get an LVAD and afterwards it was closer to 56%. So it did seem like, not only were patients making more informed decisions, but perhaps they were a little less likely to proceed with this potentially life saving therapy, but also potentially burdensome therapy.
Alex: And it's also interesting to note, I see in one of the figures here, that there was actually substantial variation across the sites in terms of reduction in implementation of an LVAD for the intervention group compared to the control group. How do you put that finding in context?
Dan: I can take that one. That's an interesting thing about the implementation design. And our implementation evaluation which we're actually in the process of writing and publishing now showed some interesting things. But there's big site variability and there's local cultures and how these sites handle LVADs. And you notice that site six, that didn't really see much change. They were also our largest site. They were in the mixed methods evaluation. There was some stuff that came out around aggressiveness and things that was interesting to note. Just different levels of how these sites do things. And each site had different challenges and issues. So we wanted to show that in the paper because I think most randomized trials see site differences like this and I think and honest way of reporting that is to actually show that and then try to understand that, rather than pretend there's not site differences.
Alex: Right, it's wonderful. And it's real world. I mean, if you got to site six, and you have heart failure, you are gonna get an LVAD. Basically, if your heart failure is bad enough.
Eric: Independent of your values.
Alex: Right. And if you go to some of these other sites, you were less likely to get an LVAD to begin with and much less likely after receiving the intervention. Those sample sizes are small caveat, caveat, caveat et cetera. But that reflects sort of the real world heterogeneity as we know that there are tremendous differences in culture of care at specific sites, but also the patients who go to one site versus another site. So it might be those different factors at play.
Dan: Yeah, that's exactly right. And just to get a little wonky we adjusted, in our statistical analyses, we adjusted for these site differences because they could've been such a powerful effect. I think it's important to note that for those who might be wondering.
Alex: So big picture, you found that decision quality improves with the decision aid intervention implemented sort of as the sites want to implement it in the real world. And that overall, as a secondary outcome, LVAD implantation decreases. Now can you talk about the linkage between those two outcomes? And that do you think that as people make better, more informed decisions fewer people will be choosing LVADs? Seems to follow, but I wonder if you have any thoughts on that area.
Dan: Ya a couple. I think there's some nuance there. I think there's a lot of literature in the shared decision making world that people often choose a less or the rate of people choosing the more "aggressive" option goes down after becoming more informed, but that's not always true. I think it's different and I think knowing what the right rate is is a little bit difficult.
Larry: That's what I was gonna say is that our goal is not to decrease utilization of LVAD necessarily. I mean it can be a life saving therapy and I think is very appropriate for carefully selected patients and whom that's kind of what they want to do. But part of the reason we chose the destination therapy population is that, that's a population who has contraindications to transplant. So they're usually older, they often have multi morbidity, or some other reason why they can't get a transplant. Often times those reasons they can't get a transplant are also reasons why they may want to be a little less aggressive with their care.
So I think it was an interesting population, first off, and it highlights that even pretty amazing therapies within the context of older patients and geriatric medicine that they may not play out exactly the same as their build. So I'm not surprised that it went down a little bit. But again, like Dan said, what the right number is, I think is a little unclear and our goal isn't just to decrease resource utilization, it's actually to help people get high value care and reduce low value care that's inconsistent with their goals.
Eric: Yeah I think it's fascinating because it feels like as a geriatrician you start seeing these people who get interventions that probably have been tested mostly in younger populations, not really in patients who require a whole lot of caregiver support for other things, are often institutionalized. And we start seeing issues when, like these LVAD patients are starting to have to think about institutionalization and nursing homes or board and care, and nobody really wants to take these people because of the liability issues of what happens if their LVAD runs out of battery. Have you guys seen that? Is that just-
Dan: We've had that problem in Denver a little bit too. A lot of nursing homes are scared of these patients because they don't know what to do with them and so they'll say they'll have to go to LTAC. And the same is true with hospice. We've done some other work looking at hospices and some hospices are comfortable with LVADs, but many are not. Largely because they don't know what to do. So when you have a destination device, I remember one of the transplant workers would always say, "Every destination device will eventually become a hospice device." And if you don't have hospices that know how to do that, then you can get yourself into trouble there too. I think that is indeed a challenge.
And just another little plug, one of our colleagues in our group, Colleen McIlvennan, one of her areas of interest is how do you help hospices take care of these patients once they get there? I think what we're finding is it requires communication, the hospices are gonna need some help from the heart failure team. You can't just send the patient to hospice and expect the hospice to know what to do. Although hospices are willing and heart failure docs are willing, they just need that conversation about it.
Eric: I think we just got our title for this podcast. Every destination device becomes a hospice device.
Dan: You know we do try to avoid that messaging though, because as we go out and we try to implement to the heart failure and cardiology world, we really are, like Larry was saying, trying to make the clear argument that we're just trying to help patients make the right decisions for them. And it's really interesting just to contrast the LVAD to the defibrillator, which we spend a lot of time working in. There’s a lot more, I would say skepticism in the world about shared decision making in defibrillators because it's a less aggressive procedure. People see it as pretty life saving. And almost every heart failure doc I've talked to about the LVAD say things like, "I would never want to put this into somebody who doesn't fully know what they're getting into."
And I think most heart failure docs they're like primary care docs, they see patients before, they see patients after. And so they see the success stories and they see the ones that don't go well at all. And they just want to make sure patients know what they're getting into. And as far as shared decision making implementation goes, this was as easy as I think it will ever get. Because the programs were so hungry to do this.
Eric: Let me ask you ... In your decision aid, are the, or somewhere along the process, are the practical aspects of where you can live with these devices, is that ever discussed? Because that is something that's incredibly important to patients and their loved ones is not just will I love longer, but where can I live? And I guess that's probably really complicated because that also depends on these micro environment settings as well.
Larry: Yeah it's a nice question. I mean we actually make kind of the medical decisions around LVAD implant including destination therapy within the existing long standing multi disciplinary conference that we call Transplant Listing Conference because so many of the issues about follow up and adherence in the case of anticoagulation for LVADs overlap with the issues we've had in the transplant population. So we have these conversations about whether we should even offer an LVAD to patients. And those conversations are not only very complicated, but they often focus on kind of social determinants of health as well as just the logistics. And we've had multiple people where ultimately they're not offered left ventricular assist device because they don't meet certain requirements, which include a dedicated caregiver. So we don't put in an LVAD in somebody who says that they're gonna be fine and they'll live in a nursing home or an LTAC. And then do it on their own. Obviously the anticoagulation issue and being able to do regular follow up is like really critical in this population because thrombosis and bleeding are really the number one cause of death after implant.
And then we have issues like we'll have patients that don't have a three prong plug in their mobile home. And until that's addressed, they're not a candidate. Which is not a usual question I think we ask for most therapies that we offer. And then what's really somewhat interesting about Denver is, Denver is a pretty major metropolitan area over two million people, but we don't have a lot of large cities nearby. So this is the only LVAD transplant center in Colorado. But there's no heart transplant center in New Mexico, Wyoming, Montana and so we draw from a large area. And so the logistics for these patients, especially older patients that have some comorbidity, it starts to factor in to whether this is a good or a bad or a not so good therapy for these patients.
So those are interesting conversations, but sometimes really tough when patients are struggling, don't want to die, but really only have this as their option.
Alex: So I want to ask you about next steps. And by way of introducing the question of next steps, read a sentence from your paper that I think introduces this point. "Despite strong efficacy data, uptake of decision aids in routine practice has been slow." And you address this in part, I think that's true, that's absolutely true that decision aids are great, but we're just not using them very often in practice. Right? And you address this in part by having this combined implementation and effectiveness trial to figure out how to implement this in real world settings.
Where do you go from here after this study? And what's the uptake then of using this decision aid outside of the study settings?
Dan: Well, so this is exciting. This is where we've written some grants to do a broad scale dissemination project across the county. We haven't heard if that's gonna be funded yet, but even short of that, while we were conducting the trial we made a decision to put it on our website. And we had over 30 programs contact us asking us to use it. Including a program in Canada that took our decision aid, the paper version, and translated it into French. And then we've learned that the country of France is not allowed to give out industry developed materials to patients, so we heard that over 75% of the LVAD programs in France are using the French version of our decision aid.
Eric: Are they flying you out there, Dan?
Dan: I'd like to. I mean I'm not trying to toot our own horn, but we're really excited about this. But it gets to the point I was making earlier that this is a space where people are actually hungry for this. I think in the shared decision making world this may be a bit of a unique example because I think the docs and the patients and everybody sort of agrees, we don't want to do this to people until they fully know what they're getting into. And anything that helps that we will embrace. Because I'd argue that LVAD is one of the most aggressive things we're doing in medicine. And for those that where it works, it works really well. At least for a few years. And so it's really their big, big trade offs.
You know there's 170 programs in America and we're hoping to be able to do some dissemination work across all of those programs and do some science around dissemination to see how many of those programs we can actually get using the tools. That's an exciting piece of this work.
Alex: That is very exciting. Anything to add, Larry?
Larry: No. I'm glad you pulled that statement out because I think that's one of the problems with academic medicine and what we do is the concept of the ivory tower, that we studied these principles that make sense within narrow context and contrived study designs, but I'm also a clinician, Dan's a clinician and we know what clinic looks like when you got a full schedule and a sick patient comes in and sets you behind. The last thing you want to be doing is spending a bunch of time using some tool that somebody told you that was a good thing to do, but doesn't really fit within your work flow and doesn't make sense.
So we have spent a lot of effort, not just designing really good tools, but designing tools that really make sense for the practical real world work flow of what we do. And we haven't mastered that, but at least we pay attention to it and we're trying to iterate in ways that make it better all the time.
Alex: Hey before we finish here, are there any other aspects of this study or implementation that you want to mention?
Larry: Well as you know medical decisions are not just about the patient and the clinician, they are about the family caregivers and the other people involved. One of the things that's a little bit unique about destination therapy LVAD is that, again at our program and many programs, you can't get an LVAD if you don't have a dedicated family caregiver. So it's a nice setting from which to think about the involvement of families and loved ones in medical decisions and the down stream consequences. So when we conducted this study not only did we prospectively enroll patients and collect questionnaires and other data from them, we did the same thing with their dedicated caregivers, who were largely spouses.
And we have a paper that should come out in the near future really detailing the findings from that work, but the kind of top line results are that, you know this is a tough decision not just for patients but also their caregivers. They share a lot of the same emotions and challenges. And at the end of the day, the caregivers are mostly very excited and committed to participating in the decisions and the after care because they want to help their loved ones. But at the same time there's this tension that exists where they're being asked to do a lot and sometimes being asked to do things they're not comfortable with.
And so recognizing that tension I think is pretty important to this kind of research. And also just to delivering this kind of clinical care. So that was a really fun part of this and we have almost 200 caregivers who participated in the study at baseline, one month and six months.
Eric: That's great. I can just imagine the anxiety component that caregivers have when you have a device that if it doesn't charge, what happens? Even if everything's working out fine, it just always wondering in the middle of your sleep, whether or not something's going to go wrong. That must be scary.
Larry: Yeah, you summarize exactly what we heard from some of them. Again, most of them really want to do it and be involved, but to recognize that that is not all butterflies and roses. That there's some tension there is important.
Eric: Go ahead, Dan.
Dan: I was gonna say, we heard from four different caregivers that they would do things like sleep in another room because they were scared to sleep next to the patient because they didn't want to accidentally kick and unplug it. But they would use things like a baby monitor to listen. Like the degree of hyper vigilance these caregivers live with is, sounds a lot like what you hear with caregivers of patients with dementia.
Eric: Yeah. Even probably more so too because it's always a threat, there must be like PTSD style symptoms in these caregivers.
Dan: Exactly. There's data on that. There absolutely is PTSD in some of these caregivers.
Eric: Well on that happy note ...
Alex: Back to going to France.
Eric: Let's talk more about having you guys go to France for some visiting professor gigs. And make sure you include us.
Eric: Because Alex can sing, so, Alex, how about end us up with a little bit more of a song.
Alex: Little bit more?
Eric: This time in French, please. So we they get invited for a visiting professor gig, we do too.
Alex: All right. [Singing]
Alex: That's such a fun song to play, Dan. Thank you for that.
Larry: It's one of my favorites too.
Dan: Is it really? That's awesome.
Larry: Oh yeah, I love that album.
Eric: Well I want to thank both Dan and Larry for joining us today. It was a great discussion about a really important topic. I want to thank all of our listeners for joining us as well. We look forward to talking with you more next week. And to all our French listeners, bonjour.
Dan: Thank you, guys.
Alex: Thank you.
Larry: Thank you.